Pith. sign in

REVIEW 3 major objections 6 minor 45 references

Phase Retrieval of Highly Strained Bragg Coherent Diffraction Patterns using Supervised Convolutional Neural Network

T0 review · 3 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A supervised CNN trained on simulated 3D diffraction patterns predicts the missing reciprocal-space phase of highly strained crystals; after inverse Fourier transform and a few hundred error-reduction iterations, this recovers objects…

desk verdict Novel reciprocal-space phase prediction with a symmetry-aware loss, but the central 'outperforms standard PR' claim lacks a controlled comparison on simulated data. read the letter →

arxiv 2507.06644 v1 pith:LZPU6QNH submitted 2025-07-09 cond-mat.mtrl-sci physics.data-an

Matteo Masto , Vincent Favre-Nicolin , Steven Leake , Clement Atlan , Marie-Ingrid Richard , Tobias Schülli , Ewen Bellec This is my paper · ORCID
classification cond-mat.mtrl-sciphysics.data-an
keywords BraggcoherentdiffractionimagingphaseretrievalconvolutionalneuralnetworkreciprocalspacehighstraindeeplearningX-raynanoparticle
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that phase retrieval, the hardest step in Bragg coherent diffraction imaging, can be solved for highly strained crystals by training a supervised convolutional network to predict the missing phase in reciprocal space. The network never sees real-space objects; it maps measured 3D diffraction intensities to phases, exploiting the structural similarity between log-scaled intensity and phase. The recovered complex amplitude is inverse-Fourier-transformed and refined with a few hundred error-reduction iterations. On simulated patterns outside the training set and on experimental patterns from two highly strained nanoparticles, the authors report reconstructions better than those from standard iterative algorithms run dozens of times. If right, the method turns a class of BCDI experiments that were often impossible into routine ones, at two to three orders of magnitude lower compute time.

What carries the argument

The load-bearing mechanism is reciprocal-space phase prediction itself, paired with the Weighted Coherent Average (WCA) loss. The loss is a single complex average, $L_{\pm}=1-\left|\frac{1}{N}\sum_k I_{\mathrm{input},k}\,\exp(i(\pm\varphi_{\mathrm{GT},k}-\varphi_{\mathrm{pred},k}))\right|$, minimized for both signs. Maximizing the modulus of that average forces all voxels' phase differences to align coherently, fixing the global offset and the wrapping ambiguity at once, while intensity weighting steers the gradient toward the strong fringes; taking the minimum over the two signs leaves the twin orientation free. The decoder has no activation on its last layer, so it can emit unbounded unwrapped phases. Together these choices let the model learn phase in the same space as the input intensity, so the skip connections of the 3D UNet transfer shared structural features directly.

What would settle it

Take a simulated or experimental high-strain pattern whose phase variation exceeds the 2π-to-5π training window, or whose particle shape is far from the trained morphologies, reconstruct it with CNN plus error reduction, and compare with a ground truth or an independent multi-peak reconstruction; if the CNN phase is systematically wrong while the constrained refinement still converges to a plausible-looking but incorrect object, the central claim fails.

Watch

Extended reading notes

Core claim

The central claim is that predicting the reciprocal-space phase directly, rather than the real-space object, is what lets deep learning succeed where iterative phasing fails for high strain. Trained on 95,000 simulated noisy 3D patterns from Wulff, Winterbottom, and random planar-cut particles with phase ranges between 2π and 5π, a UNet-style encoder-decoder with around 143 million parameters learns to output unwrapped phase maps. A custom Weighted Coherent Average loss handles three unavoidable symmetries — constant phase offset, 2π wrapping, and overall sign (the twin solution) — by comparing complex exponentials weighted by intensity and taking the better of the two sign choices. The predicted phase, combined with the measured modulus and inverse-Fourier-transformed, already gives a good support and phase estimate; refining only the border voxels with roughly 400 error-reduction iterations avoids the support-shrinking failures of conventional runs. On two experimental high-strain datasets whose conventional reconstructions, best of 60 runs, showed holes or shrunk supports, the CNN route converged to clean objects, which the paper states is the first time a deep-learning method outperforms conventional phase retrieval for this imaging technique.

Load-bearing premise

Everything rests on the simulated training set — Wulff, Winterbottom, and random planar-cut shapes with strain phases between 2π and 5π and Poisson noise — matching the real experimental data closely enough that a cropped and resized experimental pattern is effectively in-distribution for the network.

Editorial extensions

If this is right

  • High-strain BCDI datasets that standard phase retrieval cannot invert become reconstructable: the CNN provides a phase and support estimate good enough that about 400 error-reduction iterations converge to a clean object.
  • The compute drops from 20-100 independent runs of HIO/RAAR/ER, typically tens of minutes, to one forward pass plus a few hundred error-reduction iterations, a two-to-three order-of-magnitude speedup.
  • The same DL+ER recipe also improves low-strain reconstructions, with higher quality and greater reproducibility across runs.
  • Phase ranges from 2π to 5π, the regime where conventional support estimation fails, are handled by the trained model without retraining on each new pattern.
  • Because the model predicts phase in reciprocal space, any experimental pattern can be fed in after cropping and resizing to 64³ voxels, making the tool a practical front-end for routine BCDI analysis.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper does not pursue it, but the same reciprocal-space phase-prediction idea could transfer to other Fourier-domain imaging problems, such as ptychography, where intensity and phase share spatial structure; the architecture is not intrinsically tied to BCDI.
  • The border-only ER refinement is a crutch: if the CNN phase is wrong inside the object, the constrained refinement will not correct it, so out-of-distribution stress tests with strains beyond 5π or unusual particle shapes would reveal how far the trained manifold actually extends.
  • The success of the method suggests that simulation realism, not network capacity, is the likely bottleneck; richer forward models including defects, complex substrates, or multiple Bragg peaks could widen the class of samples the network can phase.
  • The WCA loss could be made adaptive by re-weighting the intensity term or adding a data-consistency check, letting the network flag its own low-confidence predictions when experimental noise or missing data deviate from the training distribution.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The manuscript proposes a supervised 3D U-Net-like CNN that predicts the reciprocal-space phase directly from log-scaled Bragg coherent diffraction intensities. The network is trained on 95,000 simulated 64^3 patterns with particle shapes drawn from Wulff, Winterbottom, and random planar-cut geometries and with simulated strain phases spanning 2π–5π. A custom loss function, the Weighted Coherent Average, is introduced to handle phase offset, wrap, and sign symmetries. The predicted phase is combined with the measured modulus, inverse Fourier transformed, and then refined with a small number of border-only Error Reduction iterations. The authors claim that this approach outperforms standard iterative phase retrieval on highly strained simulated data and on experimental data, and they report two experimental demonstrations on platinum and Pt/Pd nanoparticles.

Significance. If quantitatively established, the result would be a notable advance for BCDI: high-strain datasets are a recognized bottleneck, and a single CNN forward pass followed by a few hundred ER iterations would save substantial computation and could make previously intractable reconstructions routine. The paper has real strengths: the simulated test set is out-of-sample, the loss function explicitly addresses phase symmetries that plague end-to-end phase retrieval, the code is released, and the experimental datasets are published with DOIs. However, the abstract and conclusion assert that the model 'outperforms standard algorithms on highly strained simulated data' and 'for the first time' enables reconstructions beyond conventional PR, and these load-bearing claims are not supported by the quantitative evidence presented in the main text.

major comments (3)
  1. [§2.4, §2.5.1] The central claim that the model 'outperforms standard algorithms on highly strained simulated data' (abstract and §3) is not substantiated. Sections 2.4 and 2.5.1 show only visual central-slice comparisons for five test examples; no quantitative phase error, Fourier-shell correlation, PRTF-like metric, or head-to-head comparison with HIO/RAAR/ER on the same simulated test set is reported. Since ground truth is available for the simulated data, the authors should report, for example, the distribution of wrapped phase error inside the support, the object modulus/phase error, and the success rate of standard iterative PR on the same test patterns.
  2. [§2.5.2] The experimental demonstration does not support the outperformance claim as presented. The DL+PyNX path seeds the object with the CNN prediction and then runs 400 ER iterations updating only border voxels, which the authors state is intended to keep the final object close to the DL guess. The standard-PR baseline (400 HIO + 1000 RAAR + 300 ER, 60 runs) is compared only visually. With no ground truth for the experimental particles, a visually cleaner reconstruction is not evidence of correctness, and the constrained refinement makes the final object a lightly smoothed version of the DL prediction. The comparison should include a ground-truth-free metric such as the free log-likelihood of Ref. [47], a PRTF, or the consistency between the recovered object and the measured intensity, and the DL+ER result should be compared against an unconstrained ER or HIO refinement starting from the same initial object.
  3. [§2.1] The training distribution is restricted to simulated particles with Wulff, Winterbottom, and random planar-cut shapes and artificial strain phases limited to a 2π–5π range. The experimental strains are not independently characterized, and the experimental patterns are cropped and resized to 64^3 before prediction. The paper should provide evidence that the experimental inputs are in-distribution, or at least test sensitivity to out-of-distribution examples; without this, a plausible-looking but incorrect phase prediction could pass the constrained border-only ER refinement. A comparison of intensity statistics between the simulated and experimental data, or a perturbation study of the trained model, would help establish the generality of the method.
minor comments (6)
  1. [§2.3, Eq. (1)] Equation (1) uses I_input,k, the log-scaled BCDI intensity, but the normalization to [0,1] is described only in prose; please state explicitly whether the normalized log intensity is used in the loss and whether the same normalization is applied to experimental inputs.
  2. [§2.2] The architecture description reports about 143 million trainable parameters but does not mention regularization or normalization layers; please clarify whether dropout, batch normalization, or other regularization is used.
  3. [§2.5.2] The sentence 'the PyNX-based ER refinement typically takes 10 to 6 seconds' appears to contain a typo; presumably '10 to 60 seconds' was intended.
  4. [Figures 5 and 6] The figure captions refer to subfigures in a way that is easy to misread; please label the two experimental particles explicitly (e.g., P1 and P2) in the figures and captions so that the text 'Particle 1 (Figs. 5a-c)' and 'Particle 2 (Figs. 5 b-d)' is unambiguous.
  5. [§2.5.1] The phrase 'Barring some noise and inhomogeneities in the objects’ moduli' is vague; please quantify the modulus error or explain why modulus accuracy is secondary to phase accuracy for the strain-imaging application.
  6. [§1, References [25] and [26]] The related-work discussion would benefit from a clearer statement of whether the earlier 2D and 3D DL-based BCDI works are constrained to low strain, so that the claimed novelty for highly strained 3D experimental data is easier to evaluate.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity found; the supervised test set is an external benchmark and the minor self-citations are not load-bearing.

full rationale

The core derivation is a supervised learning pipeline: 95,000 simulated intensity-phase pairs are used for training and 3,000 for testing, so the simulated test performance is evaluated against ground-truth phases that are not used to set any model parameter. The Weighted Coherent Average loss (Eq. 1) is a training objective and not a post-hoc fit of the evaluation metric; the predicted phase is obtained by a forward pass after training on held-out data. No experimental data were used to train or tune the model, so the experimental demonstrations are external, although they lack quantitative ground-truth metrics. The ER refinement after the CNN prediction is initialized by the CNN output and is border-constrained, so the experimental 'success' is partly self-consistent with the DL guess; however, this is a validation and benchmarking weakness, not a circular derivation, because the model's simulated performance is independently checked against held-out data. The self-citations (Refs. 33, 35, 47) concern architecture blocks, shape-generation recipes, and a mode-decomposition tool; none is used as an unverified premise that defines the target result. The claim of outperforming standard algorithms on simulated data is not actually demonstrated with a head-to-head comparison, but the absence of evidence is a correctness or support concern, not circularity. Overall, the paper's central result is not equivalent to its inputs by construction, and no specific equation-level reduction from prediction to fitted parameter or self-citation chain is present.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The core of the method is a fitted function (the CNN) trained on simulations. The central claim that the model generalizes to unseen experimental patterns is an assumption about distributional overlap, not a derived result. All other ingredients (Fourier transform relation, phase symmetries, iterative refinement) are standard.

free parameters (4)
  • CNN trainable weights = ~143 million parameters
    Fitted on 95,000 simulated intensity-phase pairs; the method's predictions are a function of these weights, and no experimental ground truth was used to validate them.
  • Training phase range = 2π to 5π
    Section 2.1: phase variation within the particle is randomly chosen between 2π and 5π. This defines 'highly strained' for the model; experimental strains outside this range may not be handled.
  • Array size = 64^3 voxels
    All training and experimental data are cropped or resized to 64^3, which fixes the field of view and resolution of the prediction.
  • Number of ER refinement iterations = 400
    Section 2.5.2: 'within 400 iterations of Error Reduction'. This is a manually chosen refinement length, not an optimized parameter.
assumptions (4)
  • domain assumption Simulated diffraction patterns (PyNX forward model, Poisson noise) faithfully represent experimental BCDI data
    The network is trained entirely on simulations (Section 2.1) and applied to experimental data (Section 2.5.2). If the forward model or noise model is unrepresentative, the supervised labels are mis-specified.
  • domain assumption The training distribution of shapes and strain fields covers the experimental manifold
    Section 2.1 states the shapes are Wulff, Winterbottom, and random planar cuts and the strain phase range is 2π to 5π. Experimental particles may have different shapes, strain gradients, or magnitudes.
  • domain assumption The WCA loss and centered training data resolve the four phase symmetries (ramp, offset, wrap, sign)
    Section 2.3 relies on these symmetries being the only important ones; if additional ambiguities exist, the predicted phase could be biased even when the loss is low.
  • domain assumption Standard iterative phase retrieval converges to the correct solution from a good initial estimate
    The refinement step assumes ER iterations (Section 2.5.2) starting from the DL object reach the true solution rather than a local minimum.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Phase Retrieval of Highly Strained Bragg Coherent Diffraction Patterns using Supervised Convolutional Neural Network." pith.science (2026). https://pith.science/paper/LZPU6QNH

@misc{pith2026250706644,
  author       = {Pith},
  title        = {Pith review of: Phase Retrieval of Highly Strained Bragg Coherent Diffraction Patterns using Supervised Convolutional Neural Network},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LZPU6QNH}},
  note         = {Machine review of arXiv:2507.06644}
}
read the original abstract

In Bragg Coherent Diffraction Imaging (BCDI), Phase Retrieval of highly strained crystals is often challenging with standard iterative algorithms. This computational obstacle limits the potential of the technique as it precludes the reconstruction of physically interesting highly-strained particles. Here, we propose a novel approach to this problem using a supervised Convolutional Neural Network (CNN) trained on 3D simulated diffraction data to predict the corresponding reciprocal space phase. This method allows to fully exploit the potential of the CNN by mapping functions within the same space and leveraging structural similarities between input and output. The final object is obtained by the inverse Fourier transform of the retrieved complex diffracted amplitude and is then further refined with iterative algorithms. We demonstrate that our model outperforms standard algorithms on highly strained simulated data not included in the training set, as well as on experimental data.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 42 canonical work pages

  1. [47]

    Scientific Reports 10 (2020) https://doi

    Favre-Nicolin, V., Leake, S., Chushkin, Y.: Free log-likelihood as an unbiased metric for coherent diffraction imaging. Scientific Reports 10 (2020) https://doi. org/10.1038/s41598-020-57561-2 16

    work page doi:10.1038/s41598-020-57561-2 2020

  2. [1]

    Proceedings of the National Academy of Sciences of the United States of America 98(12), 6641–6645 (2001) https://doi.org/10.1073/pnas.111083998

    Miao, J., Hodgson, K.O., Sayre, D.: An approach to three-dimensional structures of biomolecules by using single-molecule diffraction images. Proceedings of the National Academy of Sciences of the United States of America 98(12), 6641–6645 (2001) https://doi.org/10.1073/pnas.111083998 . Accessed 2020-04-01

    work page doi:10.1073/pnas.111083998 2001

  3. [2]

    Nature Materials 8(4), 291–298 (2009) https://doi.org/10.1038/ nmat2400

    Robinson, I., Harder, R.: Coherent x-ray diffraction imaging of strain at the nanoscale. Nature Materials 8(4), 291–298 (2009) https://doi.org/10.1038/ nmat2400

    work page 2009

  4. [3]

    Journal of Applied Crystallography 55(3), 621–625 (2022) https://doi

    Richard, M.-I., Labat, S., Dupraz, M., Li, N., Bellec, E., Boesecke, P., Djazouli, H., Eymery, J., Thomas, O., Sch¨ ulli, T.U., Santala, M.K., Leake, S.J.: Bragg coherent diffraction imaging of single 20nm Pt particles at the ID01-EBS beamline of ESRF. Journal of Applied Crystallography 55(3), 621–625 (2022) https://doi. org/10.1107/S1600576722002886

    work page doi:10.1107/s1600576722002886 2022

  5. [4]

    Nature Communications 12(1), 5385 (2021) https://doi.org/10.1038/s41467-021-25625-0

    Carnis, J., Kshirsagar, A.R., Wu, L., Dupraz, M., Labat, S., Texier, M., Favre, L., Gao, L., Oropeza, F.E., Gazit, N., Almog, E., Campos, A., Micha, J.-S., Hensen, E.J.M., Leake, S.J., Sch¨ ulli, T.U., Rabkin, E., Thomas, O., Poloni, R., Hofmann, J.P., Richard, M.-I.: Twin boundary migration in an individual plat- inum nanocrystal during catalytic co oxid...

    work page doi:10.1038/s41467-021-25625-0 2021

  6. [5]

    Hofmann, F., Phillips, N.W., Das, S., Karamched, P., Hughes, G.M., Douglas, J.O., Cha, W., Liu, W.: Nanoscale imaging of the full strain tensor of specific dislocations extracted from a bulk sample. Phys. Rev. Mater. 4, 013801 (2020) https://doi.org/10.1103/PhysRevMaterials.4.013801

    work page doi:10.1103/physrevmaterials.4.013801 2020

  7. [6]

    Journal of Applied Crystallography 54, 797–802 (2021) https://doi.org/10.1107/S1600576721003113

    Godard, P.: On the use of the scattering amplitude in coherent x-ray bragg diffraction imaging. Journal of Applied Crystallography 54, 797–802 (2021) https://doi.org/10.1107/S1600576721003113

    work page doi:10.1107/s1600576721003113 2021

  8. [7]

    Nature Communications 4 (2013) https://doi.org/10.1038/ ncomms2661

    Yang, W., Huang, X., Harder, R., Clark, J.N., Robinson, I.K., Mao, H.K.: Coherent diffraction imaging of nanoscale strain evolution in a single crystal under high pressure. Nature Communications 4 (2013) https://doi.org/10.1038/ ncomms2661

    work page 2013

Show all 45 references
  1. [8]

    Estandarte, A.K.C., Lynch, C.M., Monteforte, M., Rawle, J., Nicklin, C., Robin- son, I.: Bragg coherent diffraction imaging of iron diffusion into gold nanocrystals (2018) https://doi.org/10.1088/1367-2630/aaebc1 12

    2018 doi

  2. [9]

    Catalysis Today 336, 169–173 (2019) https://doi.org/10.1016/j.cattod.2018.12.020

    Rochet, A., Suzana, A.F., Passos, A.R., Kalile, T., Berenguer, F., Santilli, C.V., Pulcinelli, S.H., Meneau, F.: In situ reactor to image catalysts at work in three- dimensions by bragg coherent x-ray diffraction. Catalysis Today 336, 169–173 (2019) https://doi.org/10.1016/j.c...

    2019 doi

  3. [10]

    Nature Communications 13(1), 3003 (2022) https://doi.org/10.1038/ s41467-022-30592-1

    Dupraz, M., Li, N., Carnis, J., Wu, L., Labat, S., Chatelier, C., Poll, R., Hof- mann, J.P., Almog, E., Leake, S.J., Watier, Y., Lazarev, S., Westermeier, F., Sprung, M., Hensen, E.J.M., Thomas, O., Rabkin, E., Richard, M.-I.: Imaging the facet surface strain state of supporte...

    2022

  4. [11]

    Nature Materials 22(6), 754–761 (2023) https://doi.org/10.1038/s41563-023-01528-x

    Atlan, C., Chatelier, C., Martens, I., Dupraz, M., Viola, A., Li, N., Gao, L., Leake, S.J., Sch¨ ulli, T.U., Eymery, J., Maillard, F., Richard, M.-I.: Imaging the strain evolution of a platinum nanoparticle under electrochemical control. Nature Materials 22(6), 754–761 (2023) ...

    2023 doi

  5. [12]

    ACS Nano 18, 13517–13527 (2024) https://doi.org/10.1021/ acsnano.3c11534

    Chatelier, C., Atlan, C., Dupraz, M., Leake, S., Li, N., Sch¨ ulli, T.U., Levi, M., Rabkin, E., Favre, L., Labat, S., Eymery, J., Richard, M.I.: Unveiling core-shell structure formation in a ni3fe nanoparticle with in situ multi-bragg coherent diffraction imaging. ACS Nano 18,...

    2024

  6. [13]

    Optik 35, 237–246 (1972)

    Gerchberg, R.W.: A practical algorithm for the determination of phase from image and diffraction plane pictures. Optik 35, 237–246 (1972)

    1972

  7. [14]

    Fienup, J.R.: Reconstruction of an object from the modulus of its fourier transform. Opt. Lett. 3(1), 27–29 (1978) https://doi.org/10.1364/OL.3.000027

    1978 doi

  8. [15]

    Review of Scientific Instruments 78(1) (2007) https://doi.org/10.1063/ 1.2403783

    Marchesini, S.: A unified evaluation of iterative projection algorithms for phase retrieval. Review of Scientific Instruments 78(1) (2007) https://doi.org/10.1063/ 1.2403783

    2007

  9. [16]

    Journal of Applied Crystallography 53, 1404– 1413 (2020) https://doi.org/10.1107/S1600576720010985

    Favre-Nicolin, V., Girard, G., Leake, S., Carnis, J., Chushkin, Y., Kieffer, J., Paleo, P., Richard, M.I.: PyNX: High-performance computing toolkit for coherent X-ray imaging based on operators. Journal of Applied Crystallography 53, 1404– 1413 (2020) https://doi.org/10.1107/S...

    2020 doi

  10. [18]

    Journal of Applied Crystal- lography 56(5), 1528–1536 (2023) https://doi.org/10.1107/S1600576723007720 https://onlinelibrary.wiley.com/doi/pdf/10.1107/S1600576723007720 13

    Zhao, J., Vartanyants, I.A., Zhang, F.: Bragg coherent modulation imaging for highly strained nanocrystals: a numerical study. Journal of Applied Crystal- lography 56(5), 1528–1536 (2023) https://doi.org/10.1107/S1600576723007720 https://onlinelibrary.wiley.com/doi/pdf/10.1107...

    2023 doi

  11. [19]

    Bragg coherent modulation imaging of highly-strained nanocrystals (submitted) (2025)

    2025

  12. [20]

    New Journal of Physics 22(1), 013021 (2020) https://doi.org/10.1088/1367-2630/ab61db

    Wang, Z., Gorobtsov, O., Singer, A.: An algorithm for bragg coherent x-ray diffractive imaging of highly strained nanocrystals. New Journal of Physics 22(1), 013021 (2020) https://doi.org/10.1088/1367-2630/ab61db

    2020 doi

  13. [21]

    Newton, M.C., Harder, R., Huang, X., Xiong, G., Robinson, I.K.: Phase retrieval of diffraction from highly strained crystals. Phys. Rev. B 82, 165436 (2010) https: //doi.org/10.1103/PhysRevB.82.165436

    2010 doi

  14. [22]

    Physical Review B - Condensed Matter and Materials Physics 78 (2008) https://doi.org/10.1103/PhysRevB.78.174110

    Minkevich, A.A., Baumbach, T., Gailhanou, M., Thomas, O.: Applicability of an iterative inversion algorithm to the diffraction patterns from inhomogeneously strained crystals. Physical Review B - Condensed Matter and Materials Physics 78 (2008) https://doi.org/10.1103/PhysRevB...

    2008 doi

  15. [23]

    Scientific Reports 8(1) (2018) https: //doi.org/10.1038/s41598-018-34525-1

    Cherukara, M.J., Nashed, Y.S.G., Harder, R.J.: Real-time coherent diffraction inversion using deep generative networks. Scientific Reports 8(1) (2018) https: //doi.org/10.1038/s41598-018-34525-1

    2018 doi

  16. [24]

    Journal of Applied Physics 128(18) (2020) https://doi.org/10.1063/5.0014725

    Scheinker, A., Pokharel, R.: Adaptive 3D convolutional neural network-based reconstruction method for 3D coherent diffraction imaging. Journal of Applied Physics 128(18) (2020) https://doi.org/10.1063/5.0014725

    2020 doi

  17. [25]

    IUCrJ 8(1), 12–21 (2021) https://doi.org/10.1107/ S2052252520013780

    Wu, L., Juhas, P., Yoo, S., Robinson, I.: Complex imaging of phase domains by deep neural networks. IUCrJ 8(1), 12–21 (2021) https://doi.org/10.1107/ S2052252520013780

    2021

  18. [26]

    npj Computational Materials 7(1) (2021) https: //doi.org/10.1038/s41524-021-00644-z

    Wu, L., Yoo, S., Suzana, A.F., Assefa, T.A., Diao, J., Harder, R.J., Cha, W., Robinson, I.K.: Three-dimensional coherent X-ray diffraction imaging via deep convolutional neural networks. npj Computational Materials 7(1) (2021) https: //doi.org/10.1038/s41524-021-00644-z

    2021 doi

  19. [27]

    npj Computational Materials 8(1) (2022) https://doi.org/10.1038/s41524-022-00803-w

    Yao, Y., Chan, H., Sankaranarayanan, S., Balaprakash, P., Harder, R.J., Cherukara, M.J.: AutoPhaseNN: unsupervised physics-aware deep learning of 3D nanoscale Bragg coherent diffraction imaging. npj Computational Materials 8(1) (2022) https://doi.org/10.1038/s41524-022-00803-w

    2022 doi

  20. [28]

    npj Computational Materials 10 (2024) https://doi.org/10.1038/ s41524-024-01208-7

    Yu, X., Wu, L., Lin, Y., Diao, J., Liu, J., Hallmann, J., Boesenberg, U., Lu, W., M¨ oller, J., Scholz, M., Zozulya, A., Madsen, A., Assefa, T., Bozin, E.S., Cao, Y., You, H., Sheyfer, D., Rosenkranz, S., Marks, S.D., Evans, P.G., Keen, D.A., He, X., Boˇ zovi´ c, I., Dean, M.P...

    2024

  21. [29]

    npj Computational Materials 9 (2023) https://doi.org/10.1038/ s41524-023-01022-7

    Maddali, S., Frazer, T.D., Delegan, N., Harmon, K.J., Sullivan, S.E., Allain, 14 M., Cha, W., Dibos, A., Poudyal, I., Kandel, S., Nashed, Y.S.G., Heremans, F.J., You, H., Cao, Y., Hruszkewycz, S.O.: Concurrent multi-peak bragg coher- ent x-ray diffraction imaging of 3d nanocry...

    2023

  22. [31]

    Mahmud, B.U., Hong, G.Y., Mamun, A.A., Ping, E.P., Wu, Q.: Deep learning- based segmentation of 3d volumetric image and microstructural analysis (2023) https://doi.org/10.3390/s23052640

    2023 doi

  23. [32]

    Xu, G., Wang, X., Wu, X., Leng, X., Xu, Y.: Development of skip connection in deep neural networks for computer vision and medical image analysis: A survey (2024)

    2024

  24. [33]

    npj Computational Materials 7(1) (2021) https://doi.org/10.1038/s41524-021-00583-9

    Lim, B., Bellec, E., Dupraz, M., Leake, S., Resta, A., Coati, A., Sprung, M., Almog, E., Rabkin, E., Schulli, T., Richard, M.I.: A convolutional neural network for defect classification in Bragg coherent X-ray diffraction. npj Computational Materials 7(1) (2021) https://doi.or...

    2021 doi

  25. [34]

    Jour- nal of Applied Crystallography 44(3), 635–640 (2011) https://doi.org/10.1107/ S0021889811009009

    Favre-Nicolin, V., Coraux, J., Richard, M.-I., Renevier, H.: Fast computation of scattering maps of nanostructures using graphical processing units. Jour- nal of Applied Crystallography 44(3), 635–640 (2011) https://doi.org/10.1107/ S0021889811009009

    2011

  26. [35]

    Journal of Applied Crystallography57(4), 966–974 (2024) https://doi.org/10.1107/S1600576724004163

    Masto, M., Favre-Nicolin, V., Leake, S., Sch¨ ulli, T., Richard, M.-I., Bellec, E.: Patching-based deep-learning model for the inpainting of Bragg coherent diffrac- tion patterns affected by detector gaps. Journal of Applied Crystallography57(4), 966–974 (2024) https://doi.org...

    2024 doi

  27. [36]

    https: //arxiv.org/abs/1606.06650

    C ¸ i¸ cek, Abdulkadir, A., Lienkamp, S.S., Brox, T., Ronneberger, O.: 3D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotation (2016). https: //arxiv.org/abs/1606.06650

    2016 arXiv

  28. [37]

    Chen, L.-C., Papandreou, G., Schroff, F., Adam, H.: Rethinking Atrous Convo- lution for Semantic Image Segmentation (2017)

    2017

  29. [38]

    In: Proceedings of the 21st ACM SIGPLAN International Conference on Functional Programming, pp

    Abadi, M.: Tensorflow: learning functions at scale. In: Proceedings of the 21st ACM SIGPLAN International Conference on Functional Programming, pp. 1–1 (2016)

    2016

  30. [39]

    Kingma, D.P., Ba, J.: Adam: A Method for Stochastic Optimization (2017)

    2017

  31. [40]

    Guizar-Sicairos, M., Fienup, J.R.: Understanding the twin-image problem in 15 phase retrieval. J. Opt. Soc. Am. A 29(11), 2367–2375 (2012) https://doi.org/10. 1364/JOSAA.29.002367

    2012

  32. [41]

    https://arxiv.org/abs/2403.15448

    Zhang, W., Wan, Y., Zhuang, Z., Sun, J.: What is Wrong with End-to-End Learning for Phase Retrieval? (2024). https://arxiv.org/abs/2403.15448

    2024 arXiv

  33. [42]

    European Synchrotron Radiation Facility

    Richard, M.-I., Bellec, E., Chatelier, C., Atlan, C., Zhao, J., Viola, A., Grimes, M., Olson, K.: Bragg coherent diffraction imaging of Pt nanoparticles on sap- phire (Version 1) [Dataset]. European Synchrotron Radiation Facility. Includes data from 0001 scan 65 (2025). https:...

    2025 doi

  34. [43]

    European Synchrotron Radiation Facility

    Richard, M.-I., Bouita, M., Leake, S., Atlan, C., Bellec, E., Olson, K., Khater, P.: Bragg coherent diffraction imaging of a PdPt nanoparticle (Ver- sion 1) [Dataset]. European Synchrotron Radiation Facility. Includes data from scan 4 (2025). https://doi.org/10.15151/ESRF-DC-2...

    2025 doi

  35. [44]

    Physical Review B - Condensed Matter and Materials Physics 68 (2003) https://doi.org/10.1103/PhysRevB.68.140101

    Marchesini, S., He, H., Chapman, N., Hau-Riege, P., Noy, A., Howells, R., Weier- stall, U., Spence, H.: X-ray image reconstruction from a diffraction pattern alone. Physical Review B - Condensed Matter and Materials Physics 68 (2003) https://doi.org/10.1103/PhysRevB.68.140101

    2003 doi

  36. [45]

    Crimmins, T.R., Fienup, J.R., Thelen, B.J.: Improved bounds on object support from autocorrelation support and application to phase retrieval. J. Opt. Soc. Am. A 7(1), 3–13 (1990) https://doi.org/10.1364/JOSAA.7.000003

    1990 doi

  37. [46]

    https://doi.org/10.5281/ zenodo.7656853

    Atlan, C.: Cdiutils: A Python Package for X-ray Bragg Coherent Diffraction Imag- ing Processing, Analysis and Visualisation Workflows. https://doi.org/10.5281/ zenodo.7656853 . https://github.com/clatlan/cdiutils

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.